Genetic Diversity and Population Structure of Rhododendron Rex Subsp

Home , Leucomeris, Rhododendron catawbiense, Rhododendron ponticum, Rhododendron rex

Article Genetic Diversity and Population Structure of Rhododendron rex Subsp. rex Inferred from Microsatellite Markers and Chloroplast DNA Sequences

1,2,3,4, 1, 1,2,3 1,2,3, Xue Zhang †, Yuan-Huan Liu †, Yue-Hua Wang and Shi-Kang Shen * 1 School of Life Sciences, Yunnan University, Kunming 650091, China; [email protected] (X.Z.); [email protected] (Y.-H.L.); [email protected] (Y.-H.W.) 2 School of Ecology and Environmental Sciences & Yunnan Key Laboratory for Plateau Mountain Ecology and Restoration of Degraded Environments, Yunnan University, Kunming 650091, China 3 Yunnan Key Laboratory of Plant Reproductive Adaptation and Evolutionary Ecology, Yunnan University, Kunming 650091, China 4 Breeding Base for State Key Laboratory of Land Degradation and Ecological Restoration in Northwest China, Key Laboratory for Restoration and Reconstruction of Degraded Ecosystem in Northwest China of Ministry of Education, Ningxia University, Yinchuan 750021, China * Correspondence: [email protected]; Tel.: +86-871-65031412 These authors contribute equals to this work. † Received: 7 February 2020; Accepted: 5 March 2020; Published: 7 March 2020

Abstract: Genetic diversity is vital to the sustainable utilization and conservation of plant species. Rhododendron rex subsp. rex Lévl. is an endangered species endemic to the southwest of China. Although the natural populations of this species are facing continuous decline due to the high frequency of anthropogenic disturbance, the genetic information of R. rex subsp. rex is not yet elucidated. In the present study, 10 pairs of microsatellite markers (nSSRs) and three pairs of chloroplast DNA (cpDNAs) were used in the elucidation of the genetic diversity, population structure, and demographic history of 11 R. rex subsp. rex populations. A total of 236 alleles and 12 haplotypes were found. A moderate genetic diversity within populations (HE = 0.540 for nSSRs, Hd = 0.788 for cpDNA markers), high historical and low contemporary gene flows, and moderate genetic differentiation (nSSR: FST = 0.165***; cpDNA: FST = 0.841***) were detected among the R. rex subsp. rex populations. Genetic and geographic distances showed significant correlation (p < 0.05) determined by the Mantel test. The species exhibited a conspicuous phylogeographical structure among the populations. Using the Bayesian skyline plot and species distribution models, we found that R. rex subsp. rex underwent a population demography contraction approximately 50,000–100,000 years ago. However, the species did not experience a recent population expansion event. Thus, habitat loss and destruction, which result in a population decline and species inbreeding depression, should be considered in the management and conservation of R. rex subsp. rex.

Keywords: Rhododendron; conservation strategies; genetic diﬀerentiation; gene ﬂow; populations contraction

1. Introduction Rhododendron is the largest woody plant genus in Ericaceae, containing more than 1000 recognized species, of which 567 species representing six subgenera are known from China [1]. Wild Rhododendron species are the major components of alpine and subalpine vegetation and widely distributed in America, Europe, and Asia, which have tropical to polar climates [2,3]. Therefore, these species are

Plants 2020, 9, 338; doi:10.3390/plants9030338 www.mdpi.com/journal/plants Plants 2020, 9, 338 2 of 16 potential genetic resources for the development of new cultivars that can adapt to diverse environmental conditions [4]. In addition, plants in the genus Rhododendron L. produce numerous chemical constituents and are recognized as an important source of bioactive phytochemicals [5]. Some Rhododendron species are used as traditional medicine in China, India, Europe, and North America against various diseases, such as inflammation, pain, skin ailments, common cold, and gastrointestinal disorders [5]. However, as an important natural resource for human daily life and ecosystem composition, most Rhododendron species are facing risk of extinction due to the high frequency of anthropogenic disturbance [6]. Thus, research on the population genetic information of Rhododendron species is undoubtedly beneficial for germplasm protection and sustainable utilization [6–9]. Inferring genetic information is recognized as the undisputed basis for the sustainable exploitation and conservation of plant diversity [10,11]. Different molecular markers are used in assessing genetic information and identifying distinct plant populations for management and conservation [12–14]. Microsatellite markers (SSRs) are used in revealing the genetic characteristics and related influence factors of plant species at individual and population levels due to their desirable advantages [13,15]. Chloroplast DNA (cpDNA), which is transmitted only through seeds in most angiosperms, is exceptionally conserved in gene content and organization, providing sufficient information for genome-wide evolutionary studies [16]. cpDNA can reveal a more highly geographical structure than a nuclear genome [17] and is generally used in the detection of phylogeographical patterns in plant species [18,19]. Thus, nSSRs and cpDNA were extensively and successfully documented to study the genetic diversity, variation, and population demographic of plant species [17,20–22]. Habitat loss and destruction are global problems that continue to threaten global biodiversity [23,24]. Firstly, habitat destruction and loss can cause a decline in the distribution range and population and limit the natural regeneration of a species. Secondly, habitat destruction and loss can increase selfing rates and decrease pollen diversity, thereby affecting a species’s reproductive success [23,25]. Finally, habitat destruction and loss increase genetic drift and inbreeding and reduce gene flow in the fragmented populations of species and substantially decrease species genetic diversity and adaptation to the changing environment. Some studies suggested that woody plants are less likely to lose genetic diversity after habitat fragmentation and destruction than herbaceous species [26]; however, recent reports showed that habitat loss and fragmentation are associated with increased level of inbreeding, reduced gene flow, genetic variation, plant progeny quality, and genetic extinction debt in woody species [24,27]. Thus, understanding the current genetic information of endangered woody plants subjected to habitat loss and destruction is necessary for effective conservation and management. Rhododendron species are not only popular woody ornamental plants but also play an important role in alpine and subalpine ecosystems. In addition, R. rex is an important wild germplasm source of the genus Rhodendron in China and an endangered plant endemic to the southwest of China [1]. Three subspecies (R. rex subsp. rex, R. rex subsp. gratum, and R. rex subsp. fictolacteum) are recognized in the R. rex complex. Recently, the wild populations of R. rex subsp. rex are facing continuous decline due to the high frequency of anthropogenic disturbance and forest destruction. Genetic information is important to the management and sustainable exploitation of species, particularly those threatened by habitat loss and destruction. However, the genetic diversity and structure of the wild populations of R. rex subsp. rex remain unexplored. In the present study, the genetic diversity and differentiation, population structure, and demographic history of 11 R. rex subsp. rex populations are inferred using 14 pairs of microsatellite markers and three cpDNA sequences. The following central questions are addressed: (1) What is the level of genetic diversity in R. rex subsp. rex? How does they apportion among/within the populations? (2) How is the genetic structure of the remnant population? Are they affected by historical and contemporary gene flows? (3) How is the phylogenetic relationship of haplotypes? Are they reflected by the demographic history in R. rex subsp. rex? This result is used to design optimum management strategies for R. rex subsp. rex conservation. Plants 2020, 9, 338 3 of 16 Plants 2020, 9, x FOR PEER REVIEW 3 of 16

2.2. Materials and Methods

2.1.2.1. Plant Material Sampling WeWe collectedcollected 212212 individualsindividuals ofof R.R. rexrex subsp.subsp. rexrex fromfrom 1111natural natural populations.populations. FourFour ofof thesethese populationspopulations (BJS,(BJS, DLT, DLT, BCL, BCL, and and JZS) JZS) with with 63 individuals 63 individuals were distributedwere distributed in Yunnan in Yunnan province, province, whereas sevenwhereas populations seven populations (QLB1, QLB2, (QLB1, QLB3, QLB2, GDX, QLB3, LJS, GDX, LZS, and LJS, YS) LZS, with and 149 YS) individuals with 149 individuals were distributed were indistributed Sichuan province, in Sichuan China province, (Table 1China). Our (Table sampling 1). Our locations sampling covered locations all the covered herbarium all samplingthe herbarium sites andsampling documented sites and sites documented of R. rex subsp. sitesrex of. DuringR. rex subsp ﬁeld sampling,. rex. During sampled field site,sampling, sampled sampled individuals, site, andsampled altitude individuals, were recorded and altitude (Figure were1 and recorded Table1). (Figure Fresh leaves 1 and wereTable collected1). Fresh fromleaves individuals were collected of R.from rex individualssubsp. rex separated of R. rex bysubsp a minimum. rex separated distance by ofa minimum 15 m and thendistance dried of in 15 silica m and gel then immediately. dried in Thesilica samplings gel immediately. were stored The atsamplings4 ◦C until were DNA stored extraction. at −4 °C until DNA extraction. −

A B C

FigureFigure 1.1.Distribution Distribution of chloroplastof chloroplast DNA DNA (cpDNA) (cpDNA) haplotypes haplotypes (A); map (A of); the map geographic of the distributiongeographic ofdistribution nuclear microsatellite of nuclear microsatellite clusters when clusters the assumed when the cluster assumed numbers cluster are numbers (B) K = 3 are and (B ()C K) K= 3= and6 in ( 11C) populationsK = 6 in 11 populations of Rhododendron of Rhododendron rex subsp. rex rex. subsp. rex.

Table 1. Details of sample locations, sample size (N), haplotype diversity (Hd), and nucleotide diversity Table 1. Details of sample locations, sample size (N), haplotype diversity (Hd), and nucleotide (Pi) surveyed for cpDNA sequences of R. rex subsp. rex. SSR—microsatellite marker. diversity (Pi) surveyed for cpDNA sequences of R. rex subsp. rex. SSR—microsatellite marker.

PopulationPopulation AltitudeAltitude N Haplotypes cpDNAcpDNA locationLocation LatitudeLatitude LongitudeLongitude N (cpDNA/SSR) Haplotypes (No.) CodeCode (m) (m) (cpDNA/SSR) (No.) HdHd PiPi BJS 24°24′31″ 100°38’15″ 2670 6/6 H1 0 0 BJS 24 24 31” 100 38’15” 2670 6/6 H1 0 0 DLT 24°28◦′570″ 100°41’47◦ ″ 2660 14/15 H1, H2, H3 0.538 0.00031 DLT 24◦28057” 100◦41’47” 2660 14/15 H1, H2, H3 0.538 0.00031 Yunnan BCL 26°3′26″ 101°03’11″ 2950 15/21 H1 0 0 Yunnan BCL 26◦3026” 101◦03’11” 2950 15/21 H1 0 0 YS 27°13′09″ 103°07’43″ 2887 16/23 H12 0 0 YS 27◦13009” 103◦07’43” 2887 16/23 H12 0 0 JZS 26°04′07″ 102°49’56″ 3250 16/21 H11 0 0 JZS 26◦04007” 102◦49’56” 3250 16/21 H11 0 0 QLB1 27°53′46″ 102°30’56″ 3250 14/22 H4, H5 0 0 QLB1 27 53 46” 102 30’56” 3250 14/22 H4, H5 0 0 QLB2 27°53◦′190″ 102°30’36◦ ″ 3303 14/23 H4, H5, H6 0.264 0.00028 QLB2 27◦53019” 102◦30’36” 3303 14/23 H4, H5, H6 0.264 0.00028 QLB3 27°54′0.4″ 102°31’44″ 3332 14/17 H4, H6 0.264 0.00028 Sichuan QLB3 27◦5400.4” 102◦31’44” 3332 14/17 H4, H6 0.264 0.00028 Sichuan GDX 28°24′29″ 103°14’33″ 2966 15/20 H7, H8, H9 0.514 0.00115 GDX 28◦24029” 103◦14’33” 2966 15/20 H7, H8, H9 0.514 0.00115 LJS 27°35′19″ 102°23’34″ 2833 15/20 H4, H5, H10 0 0 LJS 27◦35019” 102◦23’34” 2833 15/20 H4, H5, H10 0 0 LZSLZS 26°47 26◦′47480″48” 102°12’30102◦12’30”″ 33353335 16/24 16/24 H5H5 0 0 0 0 Total 11 155/212 H1–H12 0.788 0.0018 Total 11 155/212 H1–H12 0.788 0.0018

2.2. DNA Extraction, PCR Amplification, and Sequencing 2.2. DNA Extraction, PCR Amplification, and Sequencing We extracted genomic DNA of R. rex subsp. rex from the silica-dried leaves through a modified We extracted genomic DNA of R. rex subsp. rex from the silica-dried leaves through a modified cetyltrimethyl ammonium bromide (CTAB) method [28]. Purified DNA was amplified by three cetyltrimethyl ammonium bromide (CTAB) method [28]. Purified DNA was amplified by three universal cpDNA sequences (rbcL, matK, and psbA-trnH). A total of 14 SSR markers were selected universal cpDNA sequences (rbcL, matK, and psbA-trnH). A total of 14 SSR markers were selected from recently developed nuclear microsatellites in Rhododendron subg. Hymenanthes according to their clarity and reproducibility (Table S1) [29–31]. PCR amplification was performed in accordance with methods of Zhang et al. [1]. Forward SSR primers were labeled with a fluorescent dye (FAM, Plants 2020, 9, 338 4 of 16 from recently developed nuclear microsatellites in Rhododendron subg. Hymenanthes according to their clarity and reproducibility (Table S1) [29–31]. PCR amplification was performed in accordance with methods of Zhang et al. [1]. Forward SSR primers were labeled with a fluorescent dye (FAM, TAMRA, or HEX) and visualized by an ABI 3730xl automated sequencer at Sangon Biotech Services Company Ltd. (Shanghai, China). Fragment sizes were read with the GeneMapper version 4.0. CERVUS [32] was used in eliminating four loci as existing null alleles (F(Null) > 0.4) [33]. PCR products by three cpDNA intergenic spacers were sequenced in both directions by Sangon Biotech Services Company Ltd. (Shanghai, China).

2.3. Data Analysis

2.3.1. Data Analysis of Microsatellite Markers The dataset was edited and formatted with GenAlEx ver. 6.3 [34]. We used Genepop ver. 4.1.4 to test the Hardy–Weinberg equilibrium (HWE) for each locus and population [35]. The universal genetic diversity parameters were calculated using GenAlEx ver. 6.3 [34] and POPGENE ver. 1.32 [35]. Then, rarefied allelic richness (Ra), total diversity (HT), and the level of gene differentiation (GST) among R. rex subsp. rex populations were estimated by FSTAT ver. 2.9.3 [13,36]. Analysis of the molecular variance (AMOVA) was implemented in the estimation of genetic variation by using Arlequin ver. 3.11 [37,38], 3 and FST values with 10 permutations were calculated for the assessment of genetic differentiation between the pairwise populations of R. rex subsp. rex. The historical gene flow (Nm) between the pairs of R. rex subsp. rex populations was calculated using Wright’s principles using formula Nm = (1 F )/4F [39]. In addition, pollen to seed gene − ST ST flow ratio (mp/ms) was calculated using the Ennos formula [40]. To estimate contemporary migration patterns, we estimated the contemporary inter-population migrations in R. rex subsp. rex using the BayesAss version 3.0 by 3 106 Markov chain Monte Carlo (MCMC) iterations, with a burn-in of × 106 iterations and a sampling frequency of 2000 by setting delta at 0.15 (the default value) [41–43]. Isolation by distance was examined in GenAlEx ver. 6.3 on the basis of the correlation of a geographic distance for pairwise populations with F /(1 F ) value [34]. Population structure ST − ST was accessed through unweighted pair group mean analysis (UPGMA) and principal coordinate analysis. TFPGA ver. 1.3 with 5000 permutations [44] and GenAlEx ver. 6.3 [34] were used, respectively. The Bayesian clustering analysis with an admixture model to understand the population structure of R. rex subsp. rex using STRUCTURE ver. 2.2 was also explored [22,45]. K-values in the model ranged from two to 15 with 20 independent variables for each set with a burn-in of 1 105 iterations and 1 × × 105 subsequent Markov chain Monte Carlo steps [45]. The final best-fit number of the clusters was determined by ∆K values in STRUCTURE HARVESTER ver. 0.6.8 [46,47]. By performing a heterozygosity excess test, we explored the demographic history of the populations. We used two different models, namely, stepwise mutation and two-phased models, to construct the recent bottleneck statistic in BOTTLENECK ver. 1.2.02 (Sign and Wilcoxon tests) [48]. We further analyzed the genetic bottleneck with Garza–Williamson index (GWI) in Arlequin ver. 3.11 [38]. GWI lower than the critical Mc value of 0.68 indicated a reduction in population size [1,38,49].

2.3.2. Data Analysis of cpDNA Sequences We used SeqMan II [50] and Bioedit ver. 7.0.4.1 [51] to treat the raw sequencing data and manually edited and assembled these sequences [22]. Three cpDNA intergenic spacers of R. rex subsp. rex were combined using PAUP 4.0 [52]. The haplotypes and variable sites of combined cpDNA sequences were calculated by DnaSP ver. 5.0 [53]. Nei’s nucleotide diversity (Pi) and haplotype diversity (Hd) indices of R. rex subsp. rex were tested within a population and among populations. The haplotype distribution in each sampled population was plotted by ArcGIS ver. 10.2. In addition, we calculated HT and within-population gene diversity (HS) with Permut ver. 1.0 [22]. The two values of population diﬀerentiation GST and NST Plants 2020, 9, 338 5 of 16 were computed in accordance with the methods described by Pons and Petit [54] and with the same software. AMOVA of cpDNA sequences was constructed with Arlequin ver. 3.11 [37,38]. A genealogical haplotype network was constructed by Network ver. 4.2.0.1 for the estimation of the relationship per haplotype, and an indel was treated as a single mutational event [55]. The phylogenetic relationships of the as-obtained haplotypes of R. rex subsp. rex were inferred by Bayesian methods and neighbor-joining method in MrBayes ver. 3.1.2 [56]. Nerium oleander (EU916729.1, GQ997664.1 and AY899942.1) was selected as the outgroup species. The evolutionary rates of seed plants (1.01 10 9) were used for each Beast analysis in BEAST ver. × − 1.6.1 [57–59] with 107 iterations and a burn-in of 106 under the Hasegawa–Kishino–Yano (HKY) model and a strict clock. The most suitable model (HKY) was determined by Mega ver. 6.06 [60]. The results were visualized using the software FigTree ver. 1.4.2. The signatures of demographic changes in R. rex subsp. rex populations were assessed. We calculated pairwise mismatch distribution, neutrality tests (Tajima’s D and Fu’s FS), the sum of squared deviations and the raggedness index, and their p-values using DnaSP ver. 5.0 [53] and Arlequin ver. 3.11 [38].

2.3.3. Analysis of Species Distribution Model Species distribution models were constructed for the identiﬁcation of the species’ potential distribution during the last glacial maximum (LGM; ~21–18 ka) at present and in the future (model rcp45 for the years 2050, model rcp85 for the years 2070) by MAXENT v. 3.3.3k [61]. For each time period, models were run for 25 replicates, and default parameters were used. A total of 28 points comprised our 11 sampled sites and 17 records compiled in the Chinese Virtual Herbarium (www.cvh.org.cn), and 19 bioclimatic variables were obtained from the WorldClim database [62].

3. Results

3.1. SSR Data We identified 169 alleles at 10 polymorphic loci among 11 R. rex subsp. rex populations, ranging from eight (R-40, R-49) to 30 (R-30), with an average of 16.9 alleles per locus (). All loci and populations conformed to HWE (p > 0.05; Table S3). At the locus level, genetic diversity and variation exhibited certain dissimilarities. However, no remarkable difference was detected between populations (Table2). NP varied from 2 (BJS) to 12 (DLT and JZS), Ra varied from 3.071 (BJS) to 4.231 (JZS), AE varied from 2.011 (BJS) to 3.954 (YS), and I varied from 0.740 (BJS) to 1.319 (YS). The minimum values of HO (0.300) and HE (0.399) occurred in population BJS. The mean value of fixation indices (Fis = 0.171; Table2) was positive for R. rex subsp. rex populations, suggesting a slightly increased level of inbreeding.

Table 2. Genetic diversity of populations in R. rex subsp. rex.

Population NP Ra NA AE IHO HE Fis PPB (%) BCL 10 3.574 5.800 3.215 1.061 0.429 0.474 0.119 100.00% BJS 2 3.071 3.100 2.011 0.740 0.300 0.399 0.331 90.00% DLT 12 4.178 6.100 3.804 1.281 0.513 0.578 0.148 90.00% GDX 3 3.681 5.800 3.479 1.183 0.452 0.547 0.200 100.00% JZS 12 4.231 6.100 3.085 1.252 0.401 0.605 0.357 100.00% LJS 8 3.841 6.700 3.169 1.228 0.478 0.561 0.167 100.00% LZS 7 3.676 6.200 3.114 1.230 0.558 0.585 0.068 90.00% QLB1 4 3.689 5.800 3.213 1.183 0.515 0.556 0.098 100.00% QLB2 3 3.618 6.200 2.994 1.086 0.417 0.486 0.165 100.00% QLB3 5 3.718 5.900 3.118 1.187 0.498 0.541 0.111 90.00% YS 11 3.937 6.900 3.954 1.319 0.547 0.605 0.119 100.00% Mean 7 3.747 5.873 3.196 1.159 0.464 0.540 0.171 96.36%

Note: NA, mean number of alleles; AE, number of effective alleles; I, Shannon’s information index; HO, observed heterozygosity; HE, expected heterozygosity; NP, number of private alleles; Ra: rarefied allelic richness; Fis, fixation index; PPB (%), percentage of polymorphic loci. Plants 2020, 9, 338 6 of 16

AMOVA indicated that 83.53% genetic variation occurred within populations, whereas 16.47% variation was estimated among the populations (Table3). Genetic di ﬀerentiation was observed among populations (FST = 0.165, 0.15 < FST < 0.25).

Table 3. Analysis of molecular variance (AMOVA) based on 14 microsatellites and three cpDNA sequences in R. rex subsp. rex. d.f.: degrees of freedom.

Sum of Variance Percentage of Source of Variation d.f. Squares Components Variation (%) F = 0.165 SSR data Among populations 10 237.748 0.548 16.47 ST *** Within populations 413 1148.398 2.781 83.53 Total 423 1386.146 3.329 F = 0.841 cpDNA sequences Among populations 10 276.023 1.940 84.07 ST *** Within populations 144 52.919 0.367 15.93 Total 154 328.942 2.314 Note: *** p < 0.001, most signiﬁcant diﬀerence.

A high level of historical gene flow of pairwise populations was detected in R. rex subsp. rex (Table4). The minimum gene flow was generated from populations BJS and QLB2 (0.307), whereas the maximum gene flow was generated from populations QLB2 and QLB3 (7.452). The relative contribution of mp/ms was 24.775, indicating that pollen dispersal played an important role in the gene flow of R. rex subsp. rex. Except for other populations migrating to LJS, a non-significant level of inter-population contemporary migration rate between the populations of R. rex subsp. rex was detected (M < 0.05, Table5).

Table 4. Historical gene ﬂows between 11 populations of R. rex subsp. rex.

Population BCL BJS DLT GDX JZS LJS LZS QLB1 QLB2 QLB3 YS BCL 0 BJS 0.311 0 DLT 0.469 2.024 0 GDX 0.734 0.504 0.817 0 JZS 1.257 0.463 0.699 1.439 0 LJS 2.013 0.439 0.715 2.62 2.208 0 LZS 1.582 0.481 0.781 1.721 1.454 2.86 0 QLB1 1.895 0.421 0.674 1.636 2.146 10.591 2.97 0 QLB2 2.363 0.307 0.462 1.059 1.649 3.782 1.882 4.021 0 QLB3 2.029 0.365 0.572 1.093 1.582 2.785 2.099 3.461 7.452 0 YS 1.100 0.517 0.833 2.642 1.525 3.207 1.621 1.862 1.479 1.617 0

Table 5. Contemporary migration rate between populations of R. rex subsp. rex by BayesAss with 95% conﬁdence intervals.

Population-> BCL BJS DLT GDX JZS LJS LZS QLB1 QLB2 QLB3 YS BCL 0.695 0.029 0.029 0.028 0.027 0.055 0.028 0.028 0.027 0.028 0.028 BJS 0.029 0.696 0.028 0.028 0.028 0.051 0.028 0.028 0.029 0.027 0.029 DLT 0.029 0.027 0.695 0.028 0.028 0.053 0.027 0.027 0.029 0.028 0.029 GDX 0.028 0.027 0.027 0.697 0.029 0.052 0.027 0.028 0.028 0.028 0.028 JZS 0.027 0.028 0.028 0.026 0.695 0.057 0.027 0.028 0.028 0.029 0.027 LJS 0.030 0.028 0.027 0.029 0.028 0.719 0.027 0.028 0.029 0.027 0.029 LZS 0.028 0.028 0.029 0.028 0.028 0.057 0.694 0.027 0.026 0.028 0.027 QLB1 0.029 0.027 0.027 0.028 0.027 0.055 0.027 0.694 0.028 0.029 0.028 QLB2 0.029 0.028 0.026 0.029 0.027 0.055 0.028 0.029 0.695 0.027 0.027 QLB3 0.028 0.029 0.028 0.028 0.027 0.055 0.027 0.028 0.028 0.695 0.028 YS 0.028 0.028 0.026 0.028 0.027 0.056 0.028 0.026 0.029 0.028 0.695 Note: population->: population migration into the other populations. Plants 2020, 9, 338 7 of 16

The optimal K value was 3 with ∆K of 63.924, and the second ﬁt value was 6 with ∆K of 16.473 according to STUCTURE analysis (Figure S1A,B). At K of 3, the populations GDX, YS, and JZS were similar, BJS and DLT were related, and the remaining populations BCL, LJS, LZS, QLB1, QLB2, and QLB3 comprised one group (Figure1B; Figure S1A). At K = 6, the populations BCL and LZS were further distinguished from LJS, QLB1, QLB2, and QLB3, and JZS was further distinguished from GDX and YS (Figure1C; Figure S1A). This result was in accordance with the conclusions of UPGMA

(FigurePlants 2020 S1C), 9, x andFOR principalPEER REVIEW component analysis (PCA) at the population level (Figure2A). In conclusion,7 of 16 the populations of R. rex subsp. rex should be grouped into three groups according to the genetic structuregenetic structure analysis analysis results by results using by SSR using data. SSR A signiﬁcant data. A significant correlation correlation between genetic between and genetic geographic and distancesgeographic was distances determined was determined by Mantel test by (Mantelp < 0.050; test Figure (p < 0.050;2A). Figure 2A).

Figure 2. PrincipalPrincipal coordinate coordinate analysis analysis ( (AA)) and and the plot of geographicalgeographical distance against genetic distance (B) for R. rex subsp. rex by SSR datadata analysis.analysis.

As shown in Table 6 6,, mostmost ofof thethe probabilitiesprobabilities ofof thethe WilcoxonWilcoxon andand SignSign teststests underunder bothboth modelsmodels in R. rex subsp. rex populations were non-signiﬁcantnon-significant ((pp >> 0.05). In addition, the allele distribution per population was presented as a normal L-shaped distribution.distribution. The above-mentioned results indicated that the R. rex subsp.subsp. rexrex populationspopulations conformedconformed toto thethe mutation–driftmutation–drift equilibrium.equilibrium. However, GWI values werewere lowerlower thanthan the the critical criticalM Mcc indices indices (0.68), (0.68), implying implying that that the the populations populations of R.of rexR. rexsubsp. subsp.rex underwentrex underwent a demographic a demographic reduction reduction in history. in history.

Table 6. Bottleneck analysis of 11 populations in R. rex subsp. rex.

Two Phased Model (T.P.M) StepStep Mutation Mutation Model Model (S.M.M) Mode Garza–Williamson Population Two phased Model (T.P.M) (S.M.M) ModeShift Garza–WilliamsonIndex Population Sign Test Wilcoxon Test Sign Test Wilcoxon Test Wilcoxon Wilcoxon Shift Index Sign Test Sign Test BCL 0.614 0.539Test 0.170Test 0.813 L 0.336 BJS 0.211 0.410 0.068 0.545 L 0.399 BCL 0.614 0.539 0.170 0.813 L 0.336 DLT 0.399 0.652 0.183 0.839 L 0.329 BJS 0.211 0.410 0.068 0.545 L 0.399 GDX 0.158 0.862 0.002 ** 0.998 L 0.492 DLT 0.399 0.652 0.183 0.839 L 0.329 JZS 0.176 0.813 0.183 0.958 L 0.361 GDX 0.158 0.862 0.002** 0.998 L 0.492 LJS 0.178 0.862 0.169 0.958 L 0.278 JZS 0.176 0.813 0.183 0.958 L 0.361 LZS 0.074 0.947 0.003 ** 0.995 L 0.297 LJS 0.178 0.862 0.169 0.958 L 0.278 QLB1 0.370 0.423 0.181 0.947 L 0.284 LZS 0.074 0.947 0.003** 0.995 L 0.297 QLB2 0.371 0.461 0.058 0.984 L 0.323 QLB1 0.370 0.423 0.181 0.947 L 0.284 QLB3 0.065 0.862 0.074 0.984 L 0.333 QLB2 0.371 0.461 0.058 0.984 L 0.323 YS 0.612 0.461 0.389 0.722 L 0.349 QLB3 0.065 0.862 0.074 0.984 L 0.333 YS 0.612 0.461Note: ** p < 0.01,0.389 signiﬁcant di0.722ﬀerence. L 0.349

3.2. cpDNA Sequence The three cpDNAs, matK, psbA-trnH, and rbcL were 828, 398, and 658 bp in length, respectively (GenBank accession numbers: numbers: MN228019-MN228483) MN228019-MN228483).. The The 1884-bp 1884-bp combined combined cpDNA cpDNA sequences sequences of ofR. R.rex rex subsp.subsp. rexrex hadhad 18 18 polymorphic polymorphic sites sites and and 12 12 haplotypes haplotypes (H1–H12) (H1–H12) (Table 11).). TheThe detaileddetailed haplotype distribution per population is displayeddisplayed in Figure1 1.. The populations DLT (0.538 and 0.00031) and GDX (0.514 and 0.00115) exhibited additional maximum values of Hd and Pi per site, followed by QLB2 (Hd = 0.264, Pi = 0.00028) and QLB3 (Hd = 0.264, Pi = 0.00028), whereas no diversity was found in the seven remaining populations (Table 1). In summary, the total Hd and Pi for R. rex subsp. rex were 0.78768 and 0.00180, respectively. The value of HT (0.909) was higher than that of HS (0.337), and the value of NST (0. 0.774) was significantly higher than that of GST (0.629; p < 0.05; Table S4). These results indicated the remarkable phylogeographic structure among the populations of R. rex subsp. rex. AMOVA indicated that 84.07% genetic variation was partitioned among populations, whereas 15.93% was partitioned within populations (Table 3).

Plants 2020, 9, 338 8 of 16

The populations DLT (0.538 and 0.00031) and GDX (0.514 and 0.00115) exhibited additional maximum values of Hd and Pi per site, followed by QLB2 (Hd = 0.264, Pi = 0.00028) and QLB3 (Hd = 0.264, Pi = 0.00028), whereas no diversity was found in the seven remaining populations (Table1). In summary, the total Hd and Pi for R. rex subsp. rex were 0.78768 and 0.00180, respectively. The value of HT (0.909) was higher than that of HS (0.337), and the value of NST (0. 0.774) was significantly higher than that of GST (0.629; p < 0.05; Table S4). These results indicated the remarkable phylogeographic R. rex rex Plantsstructure 2020, 9 among, x FOR PEER the populations REVIEW of subsp. . AMOVA indicated that 84.07% genetic variation8 of 16 was partitioned among populations, whereas 15.93% was partitioned within populations (Table3). ThisThis resultresult waswas inconsistent inconsistent with with the resultthe ofresult nSSRs of data. nSSRs Moreover, data. significantMoreover, geneticsignificant differentiation genetic differentiationwas observed amongwas observedR. rex subsp. amongrex R.populations rex subsp. rex (F STpopulations= 0.841, p (

FigureFigure 3. 3. BayesianBayesian tree tree (A (A) )and and the the network network of of haplotypes haplotypes (B (B) )based based on on combined combined cpDNA cpDNA sequences. sequences. ((AA)) The The numbers numbers on on branches branches indicate indicate the the posterior posterior probability; probability; ( (BB) )the the size size of of the the circles circles corresponds corresponds toto the the frequency frequency of of each each haplotype, haplotype, and and the the ve verticalrtical branches branches indica indicatete mutational mutational steps. steps.

OnlyOnly the the Fu Fu and and Li’ Li’DD yieldedyielded significantly signiﬁcantly positive positive values values (p < (p 0.05;< 0.05; Table Table S5) according S5) according to the to neutralitythe neutrality test. test.This This result result indicated indicated that that no recent no recent population population expansion expansion in inR.R. rex rex subsp.subsp. rexrex occurred,occurred, andand this this was was supported supported by by the the effects eﬀects of of mismat mismatchch distributions distributions shown shown in in the the multimodal multimodal graph graph (Figure(Figure 4A).4A). Based on the Bayesian Bayesian analysis, the skyline skyline plot plot indicated indicated that that the the historical historical demographic demographic ofof R.R. rex rex subsp.subsp. rexrex populationspopulations experienced experienced a a contraction contraction event event approximately approximately 50,000–100,000 50,000–100,000 years years agoago and and had had no no recent recent expansion expansion (Figure (Figure 44B).B).

Plants 2020, 9, 338 9 of 16 Plants 2020, 9, x FOR PEER REVIEW 9 of 16

FigureFigure 4. 4. MismatchMismatch distribution distribution (A (A) )and and Bayesian Bayesian skyline skyline plot plot based based on on combined combined cpDNA cpDNA sequences sequences (B(B).). (A (A) )The The solid solid lines lines show show expected expected values, values, whereas whereas the the dashed dashed lin lineses represent represent observed observed values values underunder a amodel model of of sudden sudden population population expansion. expansion. (B (B) )The The black black line line indicates indicates effective eﬀective population population size size fluctuationﬂuctuation throughout. throughout.

3.3.3.3. Species Species Distribution Distribution Model Model AccordingAccording to to predictions predictions of of R.R. rex rex subsp.subsp. rexrex’s’s past, past, present, present, and and future future potential potential distributions, distributions, thethe predicted currentcurrent distributions distributions showed showed a clear a rangeclear contractionrange contraction relative torelative the LGM to distributions. the LGM distributions.Moreover, the Moreover, moderate habitatthe moderate suitability habitat (>0.31) su wasitability slightly (>0.31) removed was to slightly the northeastern removeddirection to the northeastern(Figure5A,B). direction The potential (Figure distribution 5A,B). The with potentia moderatel distribution to high habitat with suitability moderate ( > 0.31)to high for thehabitat years suitability2050 and 2070(>0.31) was for slightly the years extended 2050 and compared 2070 was with slightly the present-day extended compared model (Figure with 5theC,D). present-day model (Figure 5C,D).

Plants 2020, 9, 338 10 of 16 Plants 2020, 9, x FOR PEER REVIEW 10 of 16

Figure 5.5. DistributionDistribution dynamics dynamics of ofR. R. rex rexsubsp. subsp.rex usingrex using MAXENT. MAXENT. Predicted Predicted distributions distributions are shown are forshown (A) for the ( lastA) the glacial last glacial maximum maximum (LGM), (LGM), (B) the (B present,) the present, (C) the (C year) the 2050, year and2050, ( Dand) the (D year) the 2070.year Color-coded2070. Color-coded keys represent keys represent different different habitat habitat suitability. suitability. 4. Discussion 4. Discussion 4.1. Genetic Diversity in R. rex Subsp. rex Populations 4.1. Genetic Diversity in R. rex Subsp. rex Populations The genetic diversity of species reflects its long-term evolution and adaptation demographic The genetic diversity of species reflects its long-term evolution and adaptation demographic history [63]. Based on nSSR data, R. rex subsp. rex has lower genetic diversity (HE = 0.540) than the other history [63]. Based on nSSR data, R. rex subsp. rex has lower genetic diversity (HE = 0.540) than the species of Rhododendron, such as R. protistum var. giganteum (HE = 0.602) [9], R. jinggangshanicum other species of Rhododendron, such as R. protistum var. giganteum (HE = 0.602) [9], R. jinggangshanicum (HE = 0.642) [64], R. simsii (HE = 0.754) [65], R. ripense (HE = 0.800) [66], and R. brachycarpum (HE = 0.642) [64], R. simsii (HE = 0.754) [65], R. ripense (HE = 0.800) [66], and R. brachycarpum (HE = 0.815) (HE = 0.815) [67], but has higher genetic diversity than R. ferrugineum (HE = 0.500) [68]. Meanwhile, [67], but has higher genetic diversity than R. ferrugineum (HE = 0.500) [68]. Meanwhile, the genetic the genetic diversity of R. rex subsp. rex is evidently higher than that of the “narrow” species (HE = 0.420) diversity of R. rex subsp. rex is evidently higher than that of the “narrow” species (HE = 0.420) and and lower than that of the “widespread” species (HE = 0.620) [69]. Inconsistent with the results of microsatellitelower than that markers, of the the“widespread” genetic diversity species of R.(H rexE = subsp.0.620) rex[69].(Pi Inconsistent= 0.00180) assessedwith the by results cpDNA of showsmicrosatellite a higher markers, tendency the toward genetic highdiversity genetic of R. diversity rex subsp. than rex the (Pi genetic = 0.00180) diversities assessed of by 20 cpDNA species shows a higher tendency toward high genetic diversity than the genetic diversities of 20 species of of Rhododendron sect. Brachycalyx (insular species, Pi = 0.00040; continental species, Pi = 0.00160) Rhododendron sect. Brachycalyx (insular species, Pi = 0.00040; continental species, Pi = 0.00160) in East in East Asia [70], but a lower tendency than the genetic diversity of bird-dispersed arctic–alpine plant Asia [70], but a lower tendency than the genetic diversity of bird-dispersed arctic–alpine plant Vaccinium vitis-idaea in Ericaceae (Pi = 0.00240) [71]. The value of HT estimated in R. rex subsp. rex (0.909) Vaccinium vitis-idaea in Ericaceae (Pi = 0.00240) [71]. The value of HT estimated in R. rex subsp. rex is higher than the mean value of HT (0.6747) in 170 plant species according to cpDNA [19,72]. Therefore, (0.909) is higher than the mean value of HT (0.6747) in 170 plant species according to cpDNA [19,72]. R. rex subsp. rex possesses a relatively moderate genetic diversity compared with the other species Therefore, R. rex subsp. rex possesses a relatively moderate genetic diversity compared with the other in Rhododendron. In general, life span, reproductive mode, and breeding system are the important species in Rhododendron. In general, life span, reproductive mode, and breeding system are the factors in genetic diversity [6,22,69]. As in other outcrossing and long-lived species in Rhododendron, important factors in genetic diversity [6,22,69]. As in other outcrossing and long-lived species in high historical gene flow among ancestral population mitigates the loss of genetic diversity and further Rhododendron, high historical gene flow among ancestral population mitigates the loss of genetic results in a moderate or high genetic diversity in remnant populations [69,73]. Thus, the current levels diversity and further results in a moderate or high genetic diversity in remnant populations [69,73]. of genetic diversity in R. rex subsp. rex may be attributed to the species’ long-lived habit, which is Thus, the current levels of genetic diversity in R. rex subsp. rex may be attributed to the species’ long- similar to other perennial woody plants [9,22]. lived habit, which is similar to other perennial woody plants [9,22]. 4.2. Genetic Differentiation and Structure among R. rex Subsp. rex Populations 4.2. Genetic Differentiation and Structure among R. rex Subsp. rex Populations The FST value of R. rex subsp. rex indicated that a moderate genetic differentiation among populationsThe FST occurred. value of AR. total rex ofsubsp. 83.75% rex genetic indicated variation that occurreda moderate within geneticR. rex differentiationsubsp. rex populations among withpopulations regard tooccurred. nSSR markers, A total whereas of 83.7 83.53%5% genetic variation variation was partitionedoccurred within among R. populations rex subsp. with rex regardpopulations to cpDNA with sequences.regard to ThisnSSR discordance markers, wh shouldereas be83.53% affected variation by dispersal was mechanismspartitioned amongamong populations with regard to cpDNA sequences. This discordance should be affected by dispersal mechanisms among populations of plant species in Rhododendron [74,75]. Insect visitors are the

Plants 2020, 9, 338 11 of 16 populations of plant species in Rhododendron [74,75]. Insect visitors are the primary pollen dispersal vectors for Rhododendron species. Various insect vectors evolved longer dispersal distance for pollen, whereas the seeds dispersed by wind traveled less than 10 m albeit in open landscapes [74]. Meanwhile, this different consequence might be related to the type and evolutionary rates of different genome sequences [76]. In general, the evolutionary rate of nuclear genomes transmitted by parents was higher than that of maternally inherited chloroplast genomes [77]. Therefore, cpDNA variations reflected a past change, whereas nSSR variations inferred recent events in the population demographics of R. rex subsp. rex. On the basis of genetic structure analysis by SSR data, the populations of R. rex subsp. rex were grouped into three groups, and the correlation between genetic and geographic distances was significant (p < 0.05). Phylogenetic trees and genealogical haplotype networks based on cpDNA sequences showed that three reciprocally lineages were detected. This species possessed unique genetic lineages and endemic cpDNA haplotypes in its separate refuge populations. The closely related haplotypes H1, H2, and H3 were distributed in populations BJS, BCL, and DLT; H4, H5, H6, H10, and H11 occurred primarily in populations LJS, JZS, LZS, QLB1, QLB2, and QLB3; H7, H8, H9, and H12 were only detected in populations GDX and YS. Habitat dislocation and overexploitation accelerate the generation of genetic differentiation among populations [9,13]. In the sampled regions, large-scale land reclamation and unreasonable forest destruction can be observed, which resulted in habitat loss and fragmented distribution in R. rex subsp. rex natural populations. In addition, gene flow is a fundamental micro-evolutionary force, influencing genetic differentiation among populations [78,79]. The contemporary gene flow of R. rex subsp. rex is lower than that of the related species of R. protistum var. giganteum [9], which plays an important role in the formation of genetic structure and differentiation among R. rex subsp. rex populations. Moreover, breeding system is an important factor for the genetic differentiation and structure of a species [6,22,69]. Although both selfing and outcrossing are detected in Rhododendron species [80,81], the breeding system in R. rex subsp. rex is yet to be explored. Based on the positive value of fixation indices (Fis) and the phenomenon of all populations deviated from HWE in the present study, we can reasonably speculate that inbreeding is present in populations of R. rex subsp. rex. Hence, the mating system and its influences on genetic differentiation and structure must be further elucidated in R. rex subsp. rex.

4.3. Population Demographic History of the R. rex Subsp. rex Exploring the historical demography of a species can facilitate our knowledge of its ancient evolutionary environment [58]. Quaternary glaciers profoundly aﬀected the distribution and genetic variation of plant species. Tremendous global climatic oscillations during quaternary glaciations with several glacial–interglacial cycles caused the expansion and contraction of plant distribution [82]. Most plants were subjected to population demographic stability or expansion throughout the LGM [83,84]. The Bayesian skyline plot of cpDNA showed that R. rex subsp. rex experienced a notable reduction approximately 50,000–100,000 years ago. This supposition is supported by the GWI values, which are lower than the critical Mc indices (0.68). Microsatellite-based bottleneck analysis indicated that no recent bottleneck event occurred in the natural populations of R. rex subsp. rex. Therefore, the population demographic contraction detected in the R. rex subsp. rex might have been the result of climate oscillations, and the ﬁnding is consistent with the results obtained from other species, such as Cycas simplicipinna [63]. Typically, rapid population expansion occurred in the post-glacial period because temperatures increased to warm conditions [85]. However, based on neutrality and mismatch distribution tests, no recent population expansion occurred in R. rex subsp. rex. We speculate that the populations of R. rex subsp. rex might have survived in situ rather than migrating long distances to suitable habitats and that evolutionary adaptation might have occurred in the cold environment. The current existing populations of R. rex subsp. rex were limited in distribution at 2400–3400 m elevation, and this condition might partly support our speculation. Plants 2020, 9, 338 12 of 16

In addition, the complex topology of physical environmental condition in southwest China might cause geographical barriers between population migrations. This scenario was also found in the population demography of Leucomeris decora [86].

5. Conclusions The present study ﬁrstly investigated the genetic diversity, population structure, and demographic history of 11 remnant R. rex subsp. rex populations. A moderate genetic diversity, a high genetic diﬀerentiation, and a conspicuous geographical structure were detected in R. rex subsp. rex. The species possessed unique genetic lineages and endemic cpDNA haplotypes in its separate refuge populations. In addition, we found that R. rex subsp. rex experienced a population contraction approximately 50,000–100,000 years ago based on the comprehensive analysis of demographic history. Furthermore, no recent population expansion occurred in this species. Hence, the conservation of R. rex subsp. rex should focus on habitat destruction and loss, resulting in a population decline and inbreeding depression within populations. Furthermore, all the remnant adult trees of R. rex subsp. rex should receive priority protection for the maintenance of its genetic diversity. This research exhibited tremendous ecological value for the future conservation and sustainable utilization of R. rex subsp. rex and other similar plants, which are subjected to climate oscillation, inbreeding depression, overexploration, and habitat destruction.

Supplementary Materials: The following are available online at http://www.mdpi.com/2223-7747/9/3/338/s1: Figure S1. Bayesian inference of the number of clusters when K = 3 and K = 8 (A), delta K values obtained (B), and an unweighted pair-group method with arithmetic averages (UPGMA) phenogram of R. rex subsp. rex (C) based on nSSR; Table S1. The information of 14 microsatellite primers for the R. rex subsp. rex; Table S2. Summary of the 10 microsatellite loci used to the 11 populations of R. rex subsp. rex; Table S3. p-Value of Hardy–Weinberg equilibrium test for 11 populations of R. rex subsp. rex; Table S4. Genetic diversity and differentiation parameters for the combined cpDNA in 11 populations of R. rex subsp. rex; Table S5. Parameters of neutrality tests based on cpDNA of R. rex subsp. rex. Author Contributions: S.-K.S., Y.-H.L., X.Z., and Y.-H.W. initiated and designed the research. S.-K.S. obtained funding for this study. S.-K.S., Y.-H.L., and X.Z. collected the materials and performed the experiments. S.-K.S., Y.-H.L., X.Z., and Y.-H.W. wrote and revised the paper. All authors read and approved the version to be published. All authors have read and agreed to the published version of the manuscript. Funding: This study was financially supported by the National Key Research and Development Project of China (2017YFC0505204), the National Natural Science Foundation of China (31560224, 31870529), the Young Academic and Technical Leader Raising Foundation of Yunnan Province (2018HB035), the Open Fund of Yunnan Key Laboratory for Plateau Mountain Ecology and Restoration of Degraded Environments (2018DG005), and the Program for Excellent Young Talents, Yunnan University. Acknowledgments: The authors thank Xiu-Yan Feng from the Kunming Institute of Botany, Chinese Academy of Science for her construction suggestions. We also thank Fang-Li Liu, Xiong-Li Zhou, and Yue Zhang for their assistance with the field sampling. Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Zhang, Y.; Zhang, X.; Wang, Y.H.; Shen, S.K. De Novo Assembly of transcriptome and development of novel EST-SSR markers in Rhododendron rex Lévl. through illumina sequencing. Front. Plant Sci. 2017, 8, 1664. [CrossRef] 2. Yoichi, W.; Jin, X.F.; Peng, C.I.; Tamaki, I.; Tomaru, N. Contrasting diversiﬁcation history between insular and continental species of three-leaved azaleas (Rhododendron sect. Brachycalyx) in East Asia. J. Biogeogr. 2017, 44, 1065–1076. [CrossRef] 3. Xing, W.; Liao, J.; Cai, M.; Xia, Q.; Liu, Y.; Zeng, W.; Jin, X. De novo assembly of transcriptome from Rhododendron latoucheae Franch. using illumina sequencing and development of new EST-SSR markers for genetic diversity analysis in Rhododendron. Tree Genet. Genomes 2017, 13, 53. [CrossRef] 4. Wei, H.; Dhanaraj, A.L.; Rowland, L.J.; Fu, Y.; Krebs, S.L.; Arora, R. Comparative analysis of expressed sequence tags from cold-acclimated and non-acclimated leaves of Rhododendron catawbiense Michx. Planta 2005, 221, 406–416. [CrossRef] Plants 2020, 9, 338 13 of 16

5. Qiang, Y.; Zhou, B.; Gao, K. Chemical constituents of plants from the genus. Rhododendron 2011, 8, 792–815. 6. Wu, F.Q.; Shen, S.K.; Zhang, X.J.; Wang, Y.H.; Sun, W.B. Genetic diversity and population structure of an extremely endangered species: The world’s largest Rhododendron. AoB Plants 2015, 7, 10696–10700. [CrossRef][PubMed] 7. Scheiber, S.M.; Robacker, C.D.; Lindstrom, O.M. Evaluation of deciduous azaleas for cold hardiness potential in the southeastern united states. J. Environ. Hortic. 2000, 18, 230–233. 8. Scariot, V.; De, K.E.; Handa, T.; De, R.J. Comparative study of the discriminating capacity and eﬀectiveness of AFLP, STMS and EST markers in assessing genetic relationships among evergreen azaleas. Plant Breed. 2010, 126, 207–212. [CrossRef] 9. Wu, F.Q.; Shen, S.K.; Zhang, X.; Yang, G.S.; Wang, Y.H. Inferences of genetic structure and demographic history of Rhododendron protistum var. giganteum—The world’s largest Rhododendron using microsatellite markers. Flora 2017, 233, 1–6. [CrossRef] 10. Frankham, R. Genetics and extinction. Biol. Conserv. 2005, 126, 131–140. [CrossRef] 11. Divakara, B.N.; Upadhyaya, H.D.; Wani, S.P.; Gowda, C.L.L. Biology and genetic improvement of Jatropha curcas L.: A review. Appl. Energy 2010, 87, 732–742. [CrossRef] 12. Han, Y.C.; Teng, C.Z.; Chang, F.H.; Robert, G.W.; Zhou, M.Q.; Hu, Z.L.; Song, Y.C. Analyses of genetic relationships in Nelumbo nucifera using nuclear ribosomal ITS sequence data, ISSR and RAPD markers. Aquat. Bot. 2007, 87, 141–146. [CrossRef] 13. Zhang, J.J.; Ye, Q.G.; Gao, P.X.; Yao, X.H. Genetic footprints of habitat fragmentation in the extant populations of Sinojackia (Styraceaceae): Implications for conservation. Bot. J. Linn. Soc. 2012, 170, 232–242. [CrossRef] 14. Mutegi, S.M.; Muchugi, A.; Carsan, S.; Kariba, R.; Jamnadass, R.; Oballa, P.; Brunner, A.M.; Runo, S. Genetic diversity of the African poplar (Populus ilicifolia) populations in Kenya. Tree Genet. Genomes 2016, 12, 1–7. [CrossRef] 15. Wambulwa, M.C.; Meegahakumbura, M.K.; Chalo, R.; Kamunya, S.; Muchugi, A.; Xu, J.C.; Liu, J.; Li, D.Z.; Gao, L.M. Nuclear microsatellites reveal the genetic architecture and breeding history of tea germplasm of east Africa. Tree Genet. Genomes 2016, 12, 1–10. [CrossRef] 16. Pervaiz, T.; Sun, X.; Zhang, Y.; Tao, R.; Zhang, J.; Fang, J. Association between chloroplast and mitochondrial dna sequences in Chinese Prunus genotypes (Prunus persica, Prunus domestica, and Prunus avium). BMC Plant Biol. 2015, 15, 4. [CrossRef] 17. Yuan, Q.J.; Zhang, Z.Y.; Peng, H.; Ge, S. Chloroplast phylogeography of dipentodon (dipentodontaceae) in southwest China and northern Vietnam. Mol. Ecol. 2008, 17, 1054–1065. [CrossRef] 18. Hewitt, G. The genetic legacy of the Quaternary ice ages. Nature 2000, 405, 907–913. [CrossRef] 19. Petit, R.J.; Duminil, J.; Fineschi, S.; Hampe, A.; Salvini, D.; Vendramin, G.G. Comparative organization of chloroplast, mitochondrial and nuclear diversity in plant populations. Mol. Ecol. 2005, 14, 689–701. [CrossRef] 20. Garcia-Lor., A.; Luro, F.; Ollitrault, P.; Navarro, L. Genetic diversity and population structure analysis of mandarin germplasm by nuclear, chloroplastic and mitochondrial markers. Tree Genet. Genomes 2015, 11, 1–15. [CrossRef] 21. Mingeot, D.; Husson, C.; Mertens, P.; Watillon, B.; Bertin, P.; Druart, P. Genetic diversity and genetic structure of black alder (Alnus glutinosa [L.] Gaertn) in the Belgium-Luxembourg-France cross-border area. Tree Genet. Genomes 2016, 12, 1–12. [CrossRef] 22. Zhang, X.; Shen, S.K.; Wu, F.Q.; Wang, Y.H. Inferring genetic variation and demographic history of Michelia yunnanensis Franch. (Magnoliaceae) from chloroplast DNA sequences and microsatellite markers. Front. Plant Sci. 2017, 8, 583. [CrossRef][PubMed] 23. Breed, M.; Ottewell, K.; Gardner, M.; Marklund, M.H.K.; Dormontt, E.E.; Lowe, A.J. Mating patterns and pollinator mobility are critical traits in forest fragmentation genetics. Heredity 2015, 115, 108–114. [CrossRef][PubMed] 24. Aguilar, R.; Cristóbal-Pérez, E.J.; Balvino-Olvera, F.J.; Aguilar-Aguilar, M.; Aguirre-Acosta, N.; Ashworth, L.; Lobo, J.A.; Martén-Rodríguez, S.; Fuchs, E.J.; Sanchez-Montoya, G.; et al. Habitat fragmentation reduces plant progeny quality: A global synthesis. Ecol. Lett. 2019, 22, 1163–1173. [CrossRef] 25. Sáyago, R.; Quesada, M.; Aguilar, R.; Ashworth, L.; Lopezaraiza-Mikel, M.; Martén-Rodríguez, S. Consequences of habitat fragmentation on the reproductive success of two Tillandsia species with contrasting life history strategies. AoB Plants 2018, 10. [CrossRef] Plants 2020, 9, 338 14 of 16

26. Hamrick, J.L. Response of forest trees to global environmental changes. For. Ecol. Manag. 2004, 197, 323–335. [CrossRef] 27. Vranckx, G.; Jacquemyn, H.; Muys, B.; Honnay, O. Meta-analysis of susceptibility of woody plants to loss of genetic diversity through habitat fragmentation. Conserv. Biol. 2011, 26, 228–237. [CrossRef] 28. Doyle, J. DNA protocols for plants—CTAB total DNA isolation. In Molecular Techniques in Taxonomy; Hewitt, G.M., Ed.; Springer: Berlin, Germany, 1991. 29. Wang, X.; Huang, Y.; Long, C.L. Isolation and characterization of twenty-four microsatellite loci or Rhododendron decorum Franch (Ericaceae). Hortscience 2009, 44, 2028–2030. [CrossRef] 30. Wang, N.; Qin, Z.C.; Yang, J.B.; Zhang, J. Development and characterization of 15 microsatellite loci for Rhododendron delavayi Franch. (Ericaceae). Hortscience 2010, 45, 457–459. [CrossRef] 31. Li, L.F.; Yin, D.X.; Song, N. Genomic and EST microsatellites for Rhododendron aureum (Ericaceae) and cross-amplification in other congeneric species. Am. J. Bot. 2011, 98, 250–252. [CrossRef] 32. Dakin, E.E.; Avise, J.C. Microsatellite null alleles in parentage analysis. Heredity 2004, 93, 504–509. [CrossRef] [PubMed] 33. Fan, L.; Zheng, H.; Milne, R.I.; Zhang, L.; Mao, K. Strong population bottleneck and repeated demographic expansions of Populus adenopoda (salicaceae) in subtropical China. Ann. Bot. 2018, 121, 665–679. [CrossRef] [PubMed] 34. Peakall, R.; Smouse, P.E. GENALEX 6: Genetic analysis in excel. Population genetic software for teaching and research. Mol. Ecol. Notes 2006, 6, 288–295. [CrossRef] 35. Yeh, F.C.; Yang, R.; Boyle, T.B.; Ye, Z.; Mao, J.X. POPGENE, the User-Friendly Shareware for Population Genetic Analysis; University of Alberta: Edmonton, AB, Canada, 1997; Available online: http://www.ualberta.ca/ ~{}fyeh/ (accessed on 5 February 2020). 36. Goudet, J. FSTAT (Version 1.2): A computer program to calculate F-statistics. J. Hered. 1995, 86, 485–486. [CrossRef] 37. Excoffier, L.; Smouse, P.E.; Quattro, J.M. Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 1992, 131, 479–491. 38. Excoffier, L.; Laval, G.; Schneider, S. Arlequin (Version 3.0): An integrated software package for population genetics data analysis. Evol. Bioinform. 2005, 1, 47. [CrossRef] 39. Wright, S. Evolution in Mendelian populations. Genetics 1931, 16, 97. 40. Ennos, R.A. Estimating the relative rates of pollen and seed migration among plant populations. Heredity 1994, 72, 250–259. [CrossRef] 41. Abbott, R.J.; Brochmann, C. History and evolution of the arctic flora: In the footsteps of eric hultén. Mol. Ecol. 2003, 12, 299–313. [CrossRef] 42. Wilson, G.A.; Rannala, B. Bayesian inference of recent migration rates using multilocus genotypes. Genetics 2003, 163, 1177–1191. 43. Meirmans,P.G.NonconvergenceinBayesianestimationofmigrationrates.Mol.Ecol.Resour.2014,14,726–733. [CrossRef] [PubMed] 44. Miller, M.P. Tools for population genetic analyses (TFPGA) 1.3: A Windows program for the analysis of allozyme and molecular population genetic data. Comput. Softw. 1997, 4, 157. 45. Pritchard, J.K.; Stephens, M.; Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 2000, 155, 945–959. [PubMed] 46. Evanno, G.; Regnaut, S.; Goudet, J. Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol. Ecol. 2005, 14, 2611–2620. [CrossRef][PubMed] 47. Earl, D.A. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 2012, 4, 359–361. [CrossRef] 48. Piry, S.; Luikart, G.; Cornuet, J. BOTTLENECK: A computer program for detecting recent reductions in the effective size using allele frequency data. J. Hered. 1999, 90, 502–503. [CrossRef] 49. Garza, J.C.; Williamson, E.G. Detection of reduction in population size using data from microsatellite loci. Mol. Ecol. 2001, 10, 305–318. [CrossRef] 50. Swindell, S.R.; Plasterer, T.N. Sequence Data Analysis Guidebook; Humana Press: New York, NY, USA, 1997; pp. 75–89. 51. Hall, T.A. BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 1999, 41, 95–98. Plants 2020, 9, 338 15 of 16

52. Swofford, D.L. PAUP*: Phylogenetic Analysis Using Parsimony (and Other Methods); Version 4.0b10; Sinauer Associates, Inc.: Sunderland, MA, USA, 2002. 53. Librado, P.; Rozas, J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics 2009, 25, 1451–1452. [CrossRef] 54. Pons, O.; Petit, R.J. Measuring and testing genetic differentiation with ordered versus unordered alleles. Genetics 1996, 144, 1237–1245. 55. Bandelt, H.J.; Forster, P.; Röhl, A. Median-joining networks for inferring intraspecifc phylogenies. Mol. Biol. Evol. 1999, 16, 37–48. [CrossRef] [PubMed] 56. Ronquist, F.; Huelsenbeck, J.P. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 2003, 19, 1572–1574. Available online: http://tree.bio.ed.ac.uk/software/figtree/ (accessed on 5 February 2020). [CrossRef][PubMed] 57. Batalha-Filho, H.; Cabanne, G.S.; Miyaki, C.Y. Phylogeography of an Atlantic forest passerine reveals demographic stability through the last glacial maximum. Mol. Phylogenet. Evol. 2012, 65, 892–902. [CrossRef] [PubMed] 58. Graur, D.; Li, W.H. Fundamentals of Molecular Evolution; Sinauer Associates: Sunderland, MA, USA, 2000. 59. Drummond, A.J.; Rambaut, A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 2007, 7, 214. [CrossRef] 60. Tamura, K.; Stecher, G.; Peterson, D.; Filipski, A.; Kumar, S. MEGA6: Molecular evolutionary genetics analysis version 6.0. Mol. Boil. Evol. 2013, 30, 2725–2729. [CrossRef] 61. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [CrossRef] 62. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Clim. 2005, 25, 1965–1978. Available online: http://www.worldclim.org (accessed on 5 February 2020). [CrossRef] 63. Feng, X.Y.; Wang, Y.; Gong, X. Genetic diversity, genetic structure and demographic history of Cycas simplicipinna (Cycadaceae) assessed by DNA sequences and SSR markers. BMC Plant Biol. 2014, 14, 187. [CrossRef] 64. Li, M.W.; Chen, S.F.; Shi, S.; Zhang, Z.; Liao, W.B.; Wu, W.; Zhou, R.C.; Fan, Q. High genetic diversity and weak population structure of Rhododendron jinggangshanicum, a threatened endemic species in Mount Jinggangshan of China. Biochem. Syst. Ecol. 2015, 58, 178–186. [CrossRef] 65. Tan, X.X.; Li, Y.; Ge, X.J. Development and characterization of eight polymorphic microsatellites for Rhododendron simsii Planch (Ericaceae). Conserv. Genet. 2009, 10, 1553–1555. [CrossRef] 66. Kondo, T.; Nakagishi, N.; Isagi, Y. Shaping of genetic structure along Pleistocene and modern river systems in the hydrochorous riparian azalea, Rhododendron ripense (Ericaceae). Am. J. Bot. 2009, 96, 1532–1543. [CrossRef][PubMed] 67. Hirao, A.S. Kinship between parents reduces offspring fitness in a natural population of Rhododendron brachycarpum. Ann. Bot. 2010, 105, 637–646. [CrossRef] 68. Bruni, I.; De Mattia, F.; Labra, M.; Grassi, F.; Fluch, S.; Berenyi, M.; Ferrari, C. Genetic variability of relict Rhododendron ferrugineum L. populations in the Northern Apennines with some inferences for a conservation strategy. Plant Biosyst. 2012, 146, 24–32. [CrossRef] 69. Nybom, H. Comparison of different nuclear DNA markers for estimating intraspecific genetic diversity in plants. Mol. Ecol. 2004, 13, 1143–1155. [CrossRef][PubMed] 70. Yoichi, W.; Sakaguchi, S.; Ueno, S.; Tomaru, N.; Uehara, K. Development and characterization of EST-SSR markers for the genus Rhododendron section Brachycalyx (Ericaceae). Plant Species Biol. 2017, 32, 455–459. [CrossRef] 71. Ikeda, H.; Yoneta, Y.; Higashi, H.; Eidesen, P.B.; Barkalov, V.; Yakubov, V.; Brochmann, C.; Setoguchi, H. Persistent history of the bird-dispersed arctic-alpine plant Vaccinium vitis-idaea L. (Ericaceae) in Japan. J. Plant Res. 2015, 128, 437–444. [CrossRef][PubMed] 72. Hamrick, J.L.; Godt, M.J.W. Allozyme diversity in plant species. In Plant Population Genetics, Breeding, and Genetic Resources; Brown, A.H.D., Clegg, M.T., Kahler, A.L., Weir, B.S., Eds.; Sinauer: Sunderland, MA, USA, 1989; pp. 43–63. Plants 2020, 9, 338 16 of 16

73. Setoguchi, H.; Mitsui, Y.; Ikeda, H.; Nomura, N.; Tamura, A. Genetic structure of the critically endangered plant Tricyrtis ishiiana (Convallariaceae) in relict populations of Japan. Conserv. Genet. 2011, 12, 491–501. [CrossRef] 74. Escaravage, N.; Wagner, J. Pollination effectiveness and pollen dispersal in a Rhododendron ferrugineum (Ericaceae) population. Plant Biol. 2004, 6, 606–615. [CrossRef] 75. Stephenson, C.M.; Kohn, D.; Park, K.; Atkinson, R.J.; Edwards, C.; Travis, J.M.J. Testing mechanistic models of seed dispersal for the invasive Rhododendron ponticum (L.). Perspect. Plant Ecol. Evol. Syst. 2007, 9, 15–28. [CrossRef] 76. Han, Q.X.; Higashi, H.; Mitsui, Y.; Setoguchi, H. Distinct phylogeographic structures of wild radish (Raphanus sativus L. var. raphanistroides Makino) in Japan. PLoS ONE 2015, 10, e0135132. [CrossRef] 77. Cheng, B.B.; Zheng, Y.Q.; Sun, Q.W. Genetic diversity and population structure of Taxus cuspidata in the Changbai mountains assessed by chloroplast DNA sequences and microsatellite markers. Biochem. Syst. Ecol. 2015, 63, 157–164. [CrossRef] 78. Yao, X.H.; Ye, Q.G.; Kang, M.; Huang, H.W. Microsatellites analysis reveals interpopulation differention and gene flow in endangered tree Changiostyrax dolichocarpa (Styracaceae) with fragmented distribution in central China. New Phytol. 2007, 176, 472–480. [CrossRef][PubMed] 79. Côté, C.L.; Gagnaire, P.A.; Bourret, V.; Verreault, G.; Castonguay, M.; Bernatchez, L. Population genetics of the American eel (Anguilla rostrata): Fst = 0 and North Atlantic Oscillation effects on demographic fluctuations of a panmictic species. Mol. Ecol. 2013, 22, 1763–1776. [CrossRef][PubMed] 80. Escaravage, N.; Pornon, A.; Doche, B.; Till-Bottraud, I. Breeding system in an alpine species: Rhododendron ferrugineum (Ericaceae) in the French northern Alps. Can. J. Bot. 1997, 75, 736–743. [CrossRef] 81. Hahn, C.Z.; Michalski, S.G.; Durka, W. Gene flow in, and mating system of, Rhododendron simsii in a nature reserve in subtropical china. Nord. J. Bot. 2017, 35, 1–7. [CrossRef] 82. Harrison, S.P.; Yu, G.; Takahara, H.; Prentice, I.C. Diversity of temperate plants in East Asia. Nature 2001, 413, 129–130. [CrossRef] 83. Bisconti, R.; Canestrelli, D.; Colangelo, P.; Nascetti, G. Multiple lines of evidence for demographic and range expansion of a temperate species (Hyla sarda) during the last glaciation. Mol. Ecol. 2011, 20, 5313–5327. [CrossRef] 84. Pinheiro, F.; De Barros, F.; Palma-Silva, C.; Fay, M.F.; Lexer, C.; Cozzolino, S. Phylogeography and genetic differentiation along the distributional range of the orchid Epidendrum fulgens: A Neotropical coastal species not restricted to glacial refugia. J. Biogeogr. 2011, 38, 1923–1935. [CrossRef] 85. Gao, Y.D.; Zhang, Y.; Gao, X.F.; Zhu, Z.M. Pleistocene glaciations, demographic expansion and subsequent isolation promoted morphological heterogeneity: A phylogeographic study of the alpine Rosa sericea complex (Rosaceae). Sci. Rep. 2015, 5, 11698. [CrossRef] 86. Zhao, Y.J.; Gong, X. Genetic structure of the endangered Leucomeris decora (Asteraceae) in China inferred from chloroplast and nuclear DNA markers. Conserv. Genet. 2014, 13, 271–281. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).